Calibration Engineer

About the role

About us:

Calder is solving the data bottleneck for embodied AI and world models. Today's AI models lack the real-world, 4D data needed to reason on spatial problems and operate in physical environments. We capture, process, and deliver high-fidelity human-generated spatial data at scale, providing the substrate that trains the next generation of generalizable models for manipulation, navigation, and interaction.

ABOUT THE ROLE

We're looking for a Calibration Engineer to own calibration across Calder's robot and egocentric capture systems for APAC. Calibration is the core of the job: camera intrinsics and extrinsics, stereo, camera–IMU, and hand–eye transforms that downstream SLAM, reconstruction, and policy training depend on. You'll make those results repeatable by driving calibration with robot arms — programming arms to move targets and sensors through planned pose sets, so a calibration is a routine that runs the same way every time rather than a person waving a board. You'll also own enough of the robotics stack to keep that running: arm bring-up, drivers, kinematics, teleop, data collection, and playback. This is a hands-on role: you will spend time at the robot, at the calibration board, and in the stack, not only in simulation.

- Own multi-sensor calibration end to end across rigs and egocentric setups in APAC — camera intrinsics/extrinsics, stereo, camera–IMU, and camera–arm / eye-in-hand / eye-to-hand transforms

- Use robot arms to make calibration repeatable: plan and program pose trajectories that sweep targets and sensors through full coverage, automate capture and solve, and take operator-to-operator variance out of the result

- Define and hold calibration quality bars — reprojection error, coverage, observability, run-to-run repeatability — and gate captures on them

- Design and operate egocentric calibration setups (boards, poses, lighting, mounts, capture procedures) so they are repeatable and produce production-quality results

- Diagnose calibration and tracking failures from real captures — sync, coverage, board detection, extrinsics drift, kinematic error — and close the loop into procedure and software

- Stand up, maintain, and debug Calder's robotics software stack where calibration and capture depend on it: drivers, kinematics, control, teleop, logging, and playback

- Prepare robot arms, end-effectors, and lab equipment for calibration work — mounts, cameras, IMUs, compute, cabling, and safety setups

- Integrate sensors and robots into a coherent capture cell: time sync, frames, URDF/kinematic models, and calibration provenance that downstream SLAM / reconstruction / policy training can trust

- Work with hardware, firmware, and capture software on the mechanical and electrical details that affect calibration quality (camera placement, rigidity, vibration, cabling, compute)

- Write runbooks, recipes, and checklists so other operators can reproduce calibration and robot setups without tribal knowledge